Metabolites of Aspergillus terreus Antagonistic Towards the Take-all

Xifeng Li, Dahai Zhang, Uk Lee, Xianguo Li, Jiagao Cheng, Weiliang Zhu, Jee H. Jung, Hong Dae Choi, and Byeng Wha Son. Journal of Natural Products 200...
0 downloads 0 Views 206KB Size
Jmrnal OfNatural P&s Vol. 53, NO.2 , pp. 520-522, Mat-AP 1990

520

METABOLITES OF ASPERGILLUS TERREUS ANTAGONISTIC TOWARDS THE TAKE-ALL FUNGUS E.L. GHISALBERTI,~ M.J . NARBEY, and C.Y. ROWLAND Department of Organic Chemistry, The UniWJity of WertmAustralia, Nedlandc 6009, W u t m Australia

AssTRAcT.-Terrein [l]and E-4-( l-propen- l-y1)-cyclopenta- 1,2diol[2] were isolated from standing cultures of AspergihJ t-s. Evidence for the structure of 2 is presented. Both metabolites inhibit the growth of wheat take-all fungus.

Most species of Aspergillus are encountered in soil as saprophytic organisms. They are efficient soil colonizers and are frequently isolated from soil debris ( 1,2). Species of Aspergillus are known pathogens of wheat, rye grass, and other plants (1,2). As part of a project aimed at the identification of fungi in roots of wheat and rye grass, five species of Aspergillus were isolated, with Aspergillus terreus Thom (Deuteromycotina) being isolated most frequently (2). Although A . terreus was shown to kill most wheat and rye grass seedlings at high inoculum levels ( l%), with lower levels (0.5%) little or no effect was observed. Futhermore, at these low levels A . terreus reduced the root rot of wheat and rye grass caused by the take-all fungus {Gaeumannomyces graminis (Sacc.)von Arx & Oliver var tritzcz Walker (Ggt)] in sterilized and unsterilized soil. Because this isolate of A . terreus also inhibited the growth of Ggt in vitro (2) we decided to investigate the secondary metabolites produced by cultures of this fungus. We now report the isolation of terrein El] and a new compound 2 from standing cultures of A . terreus. Both metabolites show antagonism towards Ggt . Extraction of the liquid medium from standing cultures of A . terreus with EtOAc gave an extract that appeared from tlc to contain two major metabolites. Chromatography of the extract afforded, in order of decreasing polarity, the known metabolite terrein El] and a new compound 2 whose structure was deduced as follows. The compound had a molecular for-

mula C,HI40, ({MI+ 142) and showed absorption bands in its ir spectrum at 3560 and 3440 cm-', indicative of hydroxyl groups. The 13C-nmr spectrum showed only six signals, suggesting that the compound had an element of symmetry, and included signals attributable to a l-propenyl group (6, 17.66, q; 123.54, d; 135.51, d) oxymethine carbons (6, 73.06, d), methylene carbons (6, 38.59, t), and one methine carbon (6, 36.71, d). The 'H-nmr spectrum and decoupling experiments revealed the presence of an E- l-propenyl system 16, 1.62 (d, J z 4 . 8 Hz), 5.39 (dq, J = 4 . 8 , 15 Hz), 5.43 (dd, J = 6 , 15 Hz)]. The proton at 6 5.43 was coupled to an apparent sextet at 6 2.36 (J = 6 , 8 , 9.5 Hz), which showed a coupling of 8 Hz to a two-proton signal at 6 2.1 (ddd, J = 5 , 8, 13.5 Hz) and a larger coupling (9.5 Hz) to another two-proton signal at 6 1.45 (ddd,J = 5 , 9 . 5 , 13.5 Hz). Each of these sets of protons showed coupling to a signal at 6 4.0 (2H, br t , j = 5 Hz) attributed to oxymethine protons. These results can be interpreted in terms of the gross structure shown in 2 . The syn relationship of the two hydroxyl groups was established directly by formation of the acetonide 3 and indirectly from the observation that 2 was optically inactive. The relative configuration shown in 2 was deduced from nOe difference experiments, which showed small (ca. 5 % ) but significant interactions between the 3p, 5p protons (at 6 1.45, shielded by the syn C - 0 bonds) and H-l'/H-2' and between the 3 a , 5a protons (6 2.1) and the oxymethine protons (H-1 and H-2).

Mar-Apr 19901

52 1

Ghisalberti et a[. : Metabolites ofAsPergiIIus

The origin of 2 is difficult to rationalize. It could be a catabolic product of terrein which accumulates in aged cultures. Given the present evidence (3) for the biosynthesis of terrein, 2 is unlikely to be a precursor of 1.A third possibility is that 2 may arise from the proposed precursor of terrein, the lactone 4, and reflect the partitioning of 4 between two pathways (Scheme 1)whose relative importance is a function of the age of the culture. Isolates of A . twreus have been shown to be antagonistic towards the take-all fungus Ggt in vitro, to provide some protection to wheat plants from Ggt (l), and to produce metabolites that inhibit 3

O

the growth of Ggt ( 4 ) .We thus decided to test terrein and 2 for antagonism towards Ggt using a bioassay described previously (5). The results obtained (Table 1) indicate that both compounds inhibit the growth of Ggt over the shorter period of time but their effect dimminishes over 5 days. EXPERIMENTAL GENERALEXPERIMENTAL PROCEDURES.Methods, including preparation of broth culture and bioassays of compounds 1 and 2 with Ggt, are as described previously (5).

EXTRACTION AND ISOLATIoN.-Liquid medium (1.4 liters), obtained from a 3-monthold standing culture of an isolate of A . tennrr Thorn (deposited with the International

1

W

HO

$yJy 1

2

3 0

4

X=H or OH Y=HorOH

/

7

;Q-

terrein [l]

H-0

2

-

-

O

Y- C II

r OH

/

0 SCHEME1. Postulated pathway for the biosynthesis of 2.

0

Wol. 53, No. 2

Journal of Natural Products

522

Concentration (mg rn1-I)

Compoundb

1 1

2

Controld

I

Growth Period ofGgt (days at 204

3.5 2.0 2.0 -

3

5

10.3(0.5) 2 3.0 (0.8)' 2 1.0 (0.5) 29.7 (1.9)

37.0(1.2) 38.7(1.5) 34.7(0.3) 43.0 (1.2)

'Measured as diameter of colony in mm on %o strength Trypsic Soy Agar. Values are means of three replicates with SEM given in parentheses. bCompound was dissolved in 1 ml of 10% E t O W H 2 0 solution, and 40 pI of the solution was applied directly onto the inoculum plug of Ggt (9 mm). 'Mean of six replicates. d40 p1of 10% EtOH/H20 solution applied directly onto plug ofGgt (9 mm).

Mycological Institute, Kew, England, number 3 10164) (2), after filtration to remove the mycelial mat, was extracted with EtOAc. The organic layer was dried over N a W 4 and evaporated in vacuo. The crude extract (0.85 g) contained two major components (Rf0.3 and Rf 0.1) as determined by tlc {EtOAc-CH2C12 (1:4), Si gel]. These could be separated by radial plate chromatography (2 mm silica plate; CH2C12/EtOAc, gradient from 50% to 100% EtOAc). Filtration of each fraction through charcoal& gel gave compounds 1 and 2. E-44 1-propenyl-1-y1)-cyclopenta-l,2diol {2] (89 mg) was isolated as a viscous oil, R,0.3, transparent to light from 589 to 365 nm. 'H and I3C nmr are reported in text; ir (CHCI,) 3560, 3440,960 cm-'; eims m/z (I rel. int.) {MI+ 142 (24). 124 (85), 109 (loo), 83 (60), 69 106 ( l l ) , 97 (58), 96 (25), 95 (4% (42). Terrein 111(144 mg): mp 125-127'[lit. (6) mp 121-1227; ir (CHCI,) 3350, 1700, 1635 cm-'; 'H nmr (300 MHz, CDCI,) 6,6.81 (dq, ]= 15.8,6.9Hz,H-2), 6.40(brd,]= 15.8Hz, H-3), 6.02 (s, H-8), 4.89 (d,] = 2.3 Hz, H-5), 4.29(d,]=2.3Hz,H-6), 1.97(dd,]= 1.0,6.9 Hz, H-1). For spectral data obtained for Me2COd6 and D 2 0 solutions see Hill et af. (3) and Dunn et al. (6), respectively. Eims d z (% rel. int.) {MI+ 154 ( 5 ) , 139 (loo), 12 1 (43), 111(19), 109 (25), 97 ( l l ) , 95 (20), 79 (60). The mycelium was homogenized in MeOH, and the extract thus obtained was re-extracted with EtOAc. The fraction soluble in EtOAc was shown by tlc to contain further amounts of 1and 2 (0.22 gj in a 1: 1 ratio.

the acetonide 3 (30 mg), oil; m/z IM - 151+ 167; 'H m r (300 MHz, CDCI,) 6 , 5.56 (ddq,]= 15.1, 7.6, 1.4 Hz, H-1'), 5.40 (ddq,]= 15.1, 0.8, 7.1, Hz, H-27, 4.62 (m, H-1, H-2), 2.54 (apparent sextet, individual peaks of the multiplet were brcad, W,, = 4 Hz, HA), 2.0 (m, H23,H2-5), 1.64(ddd,]=7.1, 1 . 4 , 0 . 8 H ~ , H 3 3'), 1.30 and 1.26 (s, acetonide methyls); I3C n m r (75.5 MHz, CDCI,) 6, 134.7 (d, C-2'), 123.7 (d, C-1'), 111.3 (s, acetal carbon), 81.0 (d, C-1, C-2), 42.0 (d, CA), 39.0 (t, C-3, C-5), 27.0 (9)and 24.4 (9)(acetonide methyl carbons), 17.8 (q, C-3'); eims d z (% rel. int.) [M- 151+ 167 (loo), 125 (15), 107 (61), 91 (19), 85 (19), 79 (72), 68 (14), 67 (18), 59 (12), 43 (20). ACKNOWLEDGMENTS We thank Dr. K. Sivasithamparam, Soil Science and Plant Nutrition, School of Agriculture, University of Western Australia, for providing the isolate of A. tmus, and Mr. F. Almassi for carrying out the bioassay. LITERATURE CITED

1.

2. 3. 4.

FORMATIONOF ACETONIDE.-A solution of 2 (50 mg) in 2,2-dimethoxypropane (2 ml) was treated with camphor sulfonic acid (10 mg) and

5,

stirred for at room temperature' Excess K2C03was added, the solution was filtered, and the solvent was removed in vacuo. The residue in Et,O was filtered through a plug of Si gel to give

6.

K.B. Raper and D.I. Fennell, "The Genus Aspergillus," Williams and Wilkins, Baltimore, MD, 1965. M.M. Dewan and K. Sivasithamparam, A u t . ]. Bot., 36,70 1 (1988). R.A. Hill, R.H. Carter, and J. Staunton,]. C h . Soc., Perkin Tram. I , 2570(1981). C.M. Slagg and H. Fellows, ]. Agtir. Res., 75, 279 (1947). R, w, Dunlop, A. Simon, K. Sivasithamparam, and E.L. Ghisalberti, J . Nut. Prod., 52, 67 (1989). A.W. Dunn, I.D. Entwistle, and R.A.W. Johnstone, Phytochemistry, 1 4 , 2081 (1975).

Rtwiiwd 14 Alrgrrst I989